Human GMP Synthetase - The Journal of Biological Chemistry

JOURNAL OF BIOLWICAL CHEMSTRY 0 1994 by The American Society for Biochemistry and Molecular Biolom, Inc. THE

Vol. 269, No. 38, Issue of September 23, pp. 2383&23837, 1994 Printed in U.S.A.

Human GMP Synthetase PROTEIN PURIFICATION, CLONING, AND FUNCTIONAL EXPRESSION OF cDNA* (Received forpublication, April 15, 1994, and in revised form, July 11, 1994)

Margaret Hirst, Emily Haliday, John Nakamura, and Lillian Louf From the Institute of Biochemistry and Cell Biology, Syntex Discovery Research,Palo Alto, California 94303

[GMP synthetase is a key enzyme in thede nouo s y m nosuppressive therapies (11, 15, 16). In fact, in recent clinical thesis of guanine nucleotides. Human GMP synthetase trials, a potent inhibitor of IMP dehydrogenase, mycophenolic cDNA encoding acid, has shown to beeffective in the treatment of transplant has been purified to homogeneity, aand the enzyme has been isolated from the T-lymphoblasrejection and rheumatoid arthritis (16-19). The inhibition of toma cell line,A3.01. The open reading frame encodes a lymphocyte proliferation by mycophenolic acid is closely correprotein of 693 amino acids with a predicted molecular lated with the lowering of the intracellular poolof guanine weight of 76,726. The cDNA complements a guaA mutant nucleotides (10, 11, 15). Furthermore, the inhibition of prolifof Escherichia coli, which lacks a functionalGMP syn- eration by mycophenolic acid can be blocked by the additionof thetase and extracts from the transformed E. coli exhibit exogenous guanosine (10, 15, 16).These data suggest that the GMP synthetase activity, which is absent in the parental immunosuppressive activityof mycophenolic acid is a result of strain. RNA hybridization analysis shows that human the depletion of guanine nucleotide pool. Since GMP syntheGMP synthetase is encoded by a single 2.4-kilobase message. DNA hybridization analysis suggests that the hu- tase, like IMP dehydrogenase, is crucial for the de novo synman GMP synthetase is encoded by one gene. In several thesis of guanine nucleotides, the inhibitionof GMP synthetase human cell lines, the levelof mRNA expression is sub- should also result in the depletion of guanine nucleotides and, stantially higher in proliferating, transformed cellstherefore, than could be a target for immunosuppression and cancer in nontransformed cells. In two transformed cell lines, chemotherapy. Inhibiting IMP dehydrogenase and GMP syntreatment with phorbolester inhibits proliferation andthetase together maybe more potent than inhibiting each enresults in a dramatic down-regulation in the levels of zyme individually in blocking cell proliferation. For these reasons, we are interested in characterizing theGMP synthetase GMP synthetase mRNA and protein. and elucidating the mechanisms importantfor its regulation. While there is a substantial amountof published data on the In the de novo synthesis of purine nucleotides, IMP is the biochemical characterization, enzyme mechanism, cloning, and branch point metabolite at which pointthe pathwaydiverges to expression of IMP dehydrogenase(20-24), there is very limited the synthesis of either guanine or adenine nucleotides. In the information availablefor GMPsynthetase. Although the cDNA guanine nucleotide pathway,there aretwo enzymes involved in has been isolated from Escherichia coli, Bacillus subtilis, Dicconverting IMP to GMP: IMP dehydrogenase (EC 1.1.1.2051, tyostelium discoideum and Saccharomyces cerevisiae (25-281, there areno reports regarding the cloning of mammalian GMP which catalyzes the oxidation of IMP t o XMP,’ and GMP synsynthetases. There are also no reports regarding any homogethetase (EC 6.3.5.2), which catalyzes the aminationof XMP t o GMP. Accumulation of guanine nucleotides is not only essential neously purified eukaryotic GMP synthetases. Recently, we deg synthetase homogeneity to for DNA and RNA synthesis, butit also provides GTP, which is veloped a scheme for p u r i f ~ n GMP from the human T-lymphoblastoma A3.01 cell line. This puriinvolved in a number of cellular processes important for cell of fied protein has been used to study the catalytic mechanism division. GTP hydrolysis is required for microtubule assembly the humanenzyme and its interaction with inhibitors. Here we (l),protein glycosylation (21, synthesis of adenine nucleotides report thecloning and functional expressionof the humanGMP (3),protein translation (41, and activation of G proteins (5). IMP dehydrogenase and GMP synthetase are two of many synthetase gene. We demonstrate that the level of GMP synthetase is down-regulated in cell lines when proliferation is enzymes involved in cellular metabolismthat exhibit elevated inhibited. These studies provide an important foundation for levels of activity in rapidly proliferating cells, such as neoplasfurther work t o understand therole of GMP synthetase exprestic and regenerating tissues(6-8). Inhibition of IMP dehydrosion in cell growth and development. genase or GMP synthetase has been shown to result in the inhibition of cell growth (9-14). Because of the antiproliferative EXPERIMENTALPROCEDURES effect of IMP dehydrogenase and GMP synthetase inhibitors, Materials both enzymes are potential targets for anticancer and immuRestriction endonucleases and all enzymes used for plasmid construction were purchased from New England Biolabs (Beverly, MA); * The costs of publication of this article were defrayed in part by the oligo(dT)-cellulosewas from Pharmacia Biotech, Inc.; Taq DNA polympayment of page charges. Thisarticle must thereforebe hereby marked erase was from Perkin Elmer Corp.; [CY-~~PI~CTP was from Amersham “advertisement” inaccordancewith 18 U.S.C.Section 1734 solely to was from DuPont NEN.All other chemicals were Corp.; and [CY-~~PIUTP indicate this fact. purchased from Sigma. The nucleotide sequence(s) reported in this paper has been submitted to the GenBankrMIEMBL Data Bank with accession number(s) U10860. Cells and Media $ To whom correspondence should be addressed: Syntex Research, A3.01 cells were a gift from Dr. Thomas Folks (National Institute of 3401 Hillview Ave., S3-1, PaloAlto, CA 94303. Tel.: 415-852-1547;Fax: Health, Bethesda, MD) (29). The A3.01 cells were cultured in RPMI 415-354-7554. The abbreviations used are: XMP, xanthosine 5’-monophosphate; 1640 media containing5%fetal bovine serum and 10 pg/ml gentamycin IPTG, isopropyl-l-thio-a-o-galactopyranoside; PCR, polymerase chain (HyClone, Logan, UT). CEM, HL60, U937, and W138 cells were purreaction; kb, kilobase(s);TPA, 12-O-tetradecanoylphorbol-13-acetate. chased from AmericanType Culture Collection (Rockville,MD); Jurkat

23830

Alteration of Expression by Inhibition of Proliferation and UC cells weregifts from Dr. Jeffrey Northrop (Stanford University Medical Center, Stanford, CA); VB and HFF cells were providedby the cell culture facility a t Syntex. Normal peripheral T-lymphocytes were isolated from a healthy donor according to an approved protocol and purified as described (30). The bacterial hosts, SURE and XL1-Blue, used in library amplification and screening, were provided by Stratagene (La Jolla, CAI. The AT2465 strain of E. coli was obtained from the E. coli Genetic Stock Center (Yale University, New Haven, CT). The AT2465 cells have the following chromosomal markers: thi-1, guaA21, relA1, A-, and spoTl. AT2465 cells do not grow in minimal media in the absence of guanosine and thiamin (Ref. 31 and references therein). Isolation, Electrophoresis, a n d Hybridization of Blotted RNA Total cellular RNA was isolated according tothe method of Chirgwin et al. (32). Polyadenylated RNA was obtained using oligo(dT1-cellulose as described (33). RNA was electrophoresed on agarose gels in the presence of formaldehyde and transferred to Hybond N+ (Amersham Corp.) in 20 x SSPE (1x SSPE is 10 mM sodium phosphate, pH 7.7,180 mM NaCl, and 1nm EDTA) as described (34).A -300-base pair HindIIISac1 fragment from the middle of the coding regionof clone 6 was used as a probe unless specified (see Fig. 2). Prehybridization and hybridization with 32P-labeledRNA probes (1-5 x lo6 cpdml, Riboprobe system, Promega) were performed at 60 "C in 50% formamide, 5 x SSPE, 0.1% SDS, 2 x Denhardt's solution (1 x Denhardt's solution is 0.02% (w/v) each of bovine serum albumin, polyvinylpyrrolidone, and Ficoll), 0.5 mg/ml yeast RNA, and 0.1 mg/ml denatured and sheared salmon sperm DNA. The blots were washed for 1 h at 65 "C in 1 x SSPE, 1% SDS followed by a 15-min wash at 65 "C in 0.1 x SSPE and 0.1% SDS. The multiple human tissue blot was purchased from Clontech (Palo Alto, CAI. It contained poly(A)-selectedRNA fromthe following tissues: heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas. Isolation, Electrophoresis, a n d Hybridization of Blotted DNA Genomic DNA was isolated from A3.01 cells as described previously (34). Samples ofDNA (10 pg each) were digested overnight with enzymes, electrophoresed, depurinated, and denatured as described previously (34). Transfer was performed as described above for the RNA blots. Prehybridization and hybridization were carried out at 42 "C with the samebuffers, and radiolabeled probes used for RNAblots. Following hybridization, the blot was washed two times for 30 minat 50 "C in 0.2 x SSPE and 0.1% SDS. Purification of GMP Synthetasea n d Enzyme Assay Step 1: Homogenization-In a typical purification, A3.01 cells (6 x 10" cells) were lysedin 15 ml of buffer A (20 nm Tris-HC1, pH 7.6, 0.1 mM dithiothreitol, 0.5 rm EDTA, and 10%glycerol)by the use of a glass Teflon homogenizer. The homogenate was centrifuged at 15,000 x g for 20 min, and the pellet was discarded. Purification was carried out with the supernatant (cytosol). Step 2: Ammonium Sulfate Fractionation-Ammonium sulfate was added to the cytosol fraction until 35% saturation, andthe precipitated proteins were removed by centrifugation at 20,000 x g for 20 min and discarded. The proteins in the supernatant were precipitated with further addition of ammonium sulfate (60%saturation). These precipitated proteins, which included GMP synthetase, were recovered by centrifugation as above. The protein pellet was dissolved in 12 ml of buffer A. Ammonium sulfate was removed from the protein solution by gel filtration chromatography on PD-10 columns (Pharmacia). Step 3: DEAE-Cellulose Chromatography-The desalted protein fraction was applied to a 1.5 x 7.3-cm columnof DEAE-cellulose(Whatman) equilibrated in buffer A. The enzyme was eluted with a 60-ml gradient from 0 to 0.5 M sodium chloride in buffer A. The fractions containing GMP synthetase activity were pooled. Step 4: Phenyl-5PWChromatography-The DEAE-cellulose pool was diluted with an equal volume of 0.2 M potassium phosphate, pH 7.0, 1.7 M ammonium sulfate, and 20% glycerol.The protein sample was applied to a 8mm x 7.5-cm column of PhenyldPW (TosoHaas, Montgomeryville, PA) equilibrated in buffer B (0.1 M potassium phosphate, pH 7.0, 1.7 M ammonium sulfate, and 10%glycerol).GMP synthetase was eluted with a 40-ml gradient from 1.7 to 0 M ammonium sulfate in bufferB, and the fractions containing active enzyme were pooled. Step 5: Mono Q Chromatography-The Phenyl-5PW pool was desalted as above and applied to a Mono Q HR 5/5 column (Pharmacia) equilibrated with buffer C (20 mM Tris-HC1, pH 8.0,O.l rm dithiothreitol, and 20% glycerol). GMPsynthetase was eluted with a 30-ml gradient from 0 to 0.3 M sodium chloride in buffer C. The GMP synthetase

23831

fractions that contained a single band at 75 kDa on a Coomassie Bluestained SDS-polyacrylamide gel were pooled.This material is referred to as pure enzyme. Enzyme Assay during Purification The spectrophotometric-coupled assay was performed exactly as described by Spector (35). This assay measured the rate of AMP production by GMP synthetase, where the level ofAMPwas measured through the activity of three auxiliary enzymes. When crude enzyme samples, such as those in steps 1 to 3 of the purification, were examined, this assay method measured high rates ofAMP formation even in the absence of XMP. This rate is referred to as background. GMP synthetase activity is represented by the rate of AMP formation in the presence of all substrates subtracted by the rate (background)when XMP is omitted. The GMP synthetase activity determined by this assay is the same as theactivity determined by measuring radioactive GMP formation as described below (data not shown). Generation of Nucleic Acid Probe GMP synthetase from A3.01 cells was purified to homogeneity and digested with trypsin. Tryptic peptides were resolved by reverse-phase high-pressure liquid chromatography and subjected to sequence analysis by Edman degradation using an Al31470A sequencer (Applied Biosystems, Foster City, CA). Basedon the peptide sequences, degenerate oligonucleotides were synthesized using the phosphoramidite method with a Cyclone Plus DNA synthesizer (MilliGenBiosearch, Burlington, MA). Oligonucleotides weresynthesized in both the sense and antisense orientations for each peptide and used as primers in polymerase chain reactions (PCR). To generate the template, DNA complementaryto A3.01polyadenylated RNA was prepared using avian reverse transcriptase (Life Sciences, Inc.) (34). DNA amplification was performed using Thermus aquaticus (Taq) polymerase in a Perkin Elmer Corp. DNA thermal cycler (36). A specific fragment of approximately 1 kb in length was generated only when oligonucleotides 25' and SA' (corresponding to peptides 2 and 8 ) were used as primers (see Figs. 2 and 3). To confirm the identity of this fragment, it was subcloned into pCR. 1000 (Invitrogen, San Diego,CAI. From partial sequence analysis, we detected the sequences corresponding to 2s and 8A at the 5' ends of the forward and reverse strands, respectively, as well as the sequence of oligonucleotide 3s' (corresponds to peptide 3) near the 5' end of the forward strand (see Figs. 2 and 3). This PCR fragment (pcr.2S8A) was used to screen an A3.01 cDNA library. Screening of cDNA Library a n d Sequencing of Positive Clones A cDNA library, prepared by Stratagene, was constructed in the A ZAP-I1 vector system using polyadenylated RNA from A3.01cells. The resulting library yielded 9.3 x lo6 primary plaques with the host bacterial strain SURE (Stratagene). Amplification was performed on 1 x lo6 plaque-forming units. The amplified library was screened according to a standard procedure provided by Stratagene. The PCR fragment pcr.2S8A wasradiolabeled by random hexamer priming and used as a probe. A totalof 0.9 x lo6 plaque-forming units were screened. After three rounds of screening, eight positive phagemids were excisedand rescued by R407 helper phage (Stratagene). Double-stranded pBluescript (SK-) plasmids were isolated, and four of the longest clones wereanalyzed further by restriction mapping and found to be overlapping. The cDNA inserts of the four positive clones were sequenced by the method of Sanger using fluorescent dye-labeled terminators with a 373A DNA Sequencer (Applied Biosystems).Partial sequence analysis confirmedthat the four clones were overlapping.The complete sequence of both strands of the insert of clone 6 (GMPS.6) wasdetermined. Construction of Expression Plasmids andComplementation in AT2465 Cells The insert of GMPS.6 was excised frompBluescript and cloned into the NcoI-PstIcloning site of the prokaryotic expression vector, pTRC99A(Pharmacia),to yield vector TRC.6.Using TRC.6, twoexpression vectors,NF.6 and ES.6, wereconstructed to remove the 5"untranslated sequence of the cDNA. A 201-base pair fragment was removed from the 5' end of the TRC.6 by digestion with NcoI and MscI and replaced with a 60-base pair oligonucleotide that had the identical sequence of residues -1 to 59 except fora G to C change at residue -1

' Sequence of oligonucleotides: 2S,5'-AA(CfI')CA(A/G)GA(A/G)CA(A/ G)GT(A/GICPT)AT(AICPT)GC(A/GfCPT)GT-3': 8A.5'-GC(A/GICPT)GG(AI

23832

Alteration of Expression by lnhibition of Proliferation

(see Fig. 3, residue 1 represents the start site of translation). This base change convertedthe BglI site at the start of site translation to an NcoI site. The 60-base pair oligonucleotide which has the sequence 5'4ATGGC... wasdirectlyligatedinto the NcoI/MscI cut vector, and this expression vector was designated NF.6 (NF abbreviates no fusion). A second expression vector, ES.6, was constructed in the same manner, except that two stopcodons were inserted in the ligated oligonucleotide 39 base pairs downstreamfrom the start siteof translation (ES abbreviates early stop). The nucleotide sequence from residues 40 to 45 was changed from . . . G GAGGA... to . . . TG A m . . . . To test for complementation of the guanosine requirement, AT2465 cells that were transformed with NF.6 or ES.6 were spread on M9 minimal medium plates with or without guanosine (100 pg/ml). All plates were supplemented with 0.4% glucose, 0.29 casamino acids, 20 pg/ml IPTG and 100 pg/ml of the following: thiamin, glutamine, histidine, arginine, inosine, biotin, 2'-deoxyuridine, and ampicillin. To determine GMP synthetase activity, single colonies were selected and inoculated inLB media containing 100 pg/ml ampicillin. Cultures were allowed to grow to mid-log phase, and thenIPTG was added to a final concentration of 20 pg/ml. Cells were harvested3 h after the addition of IPTG.

Determination of Enzyme Activity in nansformed E. coli Cells Cells were resuspended in lysis buffer (20 mhq Tris-HCI, pH 7.6, 0.5 mM dithiothreitol, 0.2 mg/ml lysozyme, 2 mM phenylmethylsulfonyl fluoride, 0.5 mM EDTA, and 10%glycerol) and lysed a t 37 "C for 10 min. Genomic DNA was sheared by passing the lysate through a 27 gauge needle three to four times. cytosolic The fraction was separatedfrom the membrane fraction by centrifugation a t 15,000 x g for 15 min. GMP synthetase activity in the cytosol was determined by measuring the formation of [8-'"C]GMP from [8-'4C1XMP (50 mCi/mmol, Moravek, Brea, CAI. The assay mixture contained 75 mM "is-HCI, pH 7.8,lO mM MgCI,, 2 mM ATP, 1mM ['"ClXMP (10 mCi/mmolI, 5 mhf glutamine and 50 mM dithiothreitol. Typically, 5 pl of cytosol was added to 25 p1 of assay mix, and thereaction was allowed to proceed a t 40 "C for 15 min. The reaction was stopped by the addition of 6 p1 of quench solution, which contained 250 mxq EDTA and 62.5 mM each of GMP andXMP a s carriers. GMP andXMP were separated by thin layer chromatography as described (37). Briefly, the quenched reaction mixtures (IO pl) were streaked onto 2.5 cm x 20 cm strips of polyethyleneimine cellulose plates (Macherey-Nagel, Germany). The strips were developed in 2 M formic acid. Theposition of GMP was visualized by ultraviolet light and marked. The GMP band was excised, and the amount of ["CIGMP was determined by liquid scintillation counting. Protein concentration was determined by Bradford analysis (Bio-Rad).

1

97 kDa

-

66 kDa

-

45 kDa

-

2

3

4

+GMP synthetase

FIG.1. Purification of G M P synthetase from A3.01 cells. Samples from each step of the purification were subjectedto SDSpolyacrylamide gel electrophoresis and stained with Coomassie Blue. The purification step and amountof protein loaded in each lane are as follows: lane I , 35-609 ammonium sulfate precipitate, 6 pg; lane 2, DEAE-cellulose pool, 2.5 pg; lane 3 , Phenyl-5PW pool, 3 pg; lane 4, Mono Q pool, 3 pg. RESULTS

Zsolation of Clones and Sequence Analysis-In order toclone the cDNA for human GMP synthetase, theenzyme was purified more than 900-fold t o homogeneity from A3.01 cells (Fig. 1and Table I ) , and tryptic peptides from the purified protein were generated. The sequences of nine peptides were determined (Table ID, and, based on these peptide sequences, oligonucleotides in both the sense and antisense orientations were made as PCR primers. Sincethe location of the peptides inthe protein sequencewasunknown,each oligonucleotide was combined with eachof the others primers. as A fragment of approximately 1kb in length was generated usingoligonucleotides corresponding topeptides 2and 8 (Figs. 2and 3). This fragment was specific to 2 and 8 and was not generated in the absence of either oliGeneration of Anti-GMP Synthetase Antibody The HindIII-Sac1 fragment used for RNA blots was subcloned into gonucleotide or when oligonucleotides complementary to 2 and the bacterial expression vector pGEX-2T (Pharmacia) and expressed as 8 were used in thereaction instead (datanot shown).The idena glutathione transferase fusion protein in SURE bacteria. This fusion tity of this fragment, pcr.2S8A, was confirmed by sequence protein was isolated using a glutathione-Sepharose column according to analysis. Pharmacia instructions. A 1-liter culture yielded 1.5 mg of protein, Using pcr.2S8A as a probe, we screened a A ZAF"I1 cDNA which was used to immunize rabbits accordingantoapproved protocol. Purified human GMP synthetase was used in immunoblot analysis to library prepared from polyadenylated RNAof A3.01 cells. A number of positive clones were isolated of which four of the test for antibody production. longest were examined and found to be overlapping. The longImmunoprecipitation and Immunoblot Analysis est clone, GMPS.6, was approximately 2.2 kb (Fig. 2). Cells were lysed in a buffer consisting of 25 mM Hepes, pH 7.3, 150 Sequence analysis of GMPS.6, revealed an open reading mht NaCI, 1% Nonidet P-40, 10 pg/ml leupeptin, and 0.5 mM phenyl- frame of 2079 base pairs (Fig. 3). The firstATG was found 142 methylsulfonyl fluoride. Lysates (600 pg of protein each) were immubases downstream from the 5' end, anda stop codon, TAA, was noprecipitated a t 5 "C for 18 h with 20 p1 of GMP synthetase antisera and 75pl of 50% protein A-Sepharose (final volume, 500 pl). As controls, found 11bases upstream from the 3' end. No poly(A) tail was found on any of the clones. The open reading frame yielded a identical samples were treated in parallel with preimmune sera. The immunoprecipitated proteins were separated by SDS-polyacrylamide protein of 693 amino acid residues with a predicted molecular gel electrophoresis and transferred to Immobilon-P (Millipore) memweight of 76,725. The predicted size of the enzyme is in agreebranes. The membrane blots were incubated withthe GMP synthetase ment with the size of the purifiedA3.01 GMP synthetase, which antisera (1:500 dilution) for 18 h at room temperature, followed by stands at75 kDa,as indicated by SDS-polyacrylamide gel elecincubationwithperoxidase-conjugatedproteinG(Calbiochem).The trophoresis (Fig. 1).The sequences corresponding to all nine protein bands of interest were then visualized using the enhanced tryptic peptides were located within the predicted protein sechemiluminescence (ECL) detection system (Amersham Corp.). quence (Fig. 3). Phorbol Ester Deatment of Cells Sequence Similarities-Using computer analysis (Genetics U937 and HL60cells were inoculateda t a densityof 0.5 x lo6 cells/ml Computer Group,W I ) ,the predicted amino acid sequence of the 20 ng/ml of 12-0-tetradecanoylphorbol-13-acetate andtreatedwith human GMP synthetase was compared with the published se(TPAI. Control cells were treated with vehicle (ethanol) in parallel. quences of Dictyostelium and E. coli GMP synthetases. The Three days later, thecells were harvested and divided. One half of the cell pellet was used to prepare RNA for Northern analysis, and the other protein sequence of GMP synthetase is well conserved across species. The predicted protein sequence of GMPS.6 is 51.7 and half was used to prepare cell extracts for immunoprecipitation.

Alteration of Expression by Inhibition of Proliferation TABLE I Purification of GMP synthetase fromA3.01 cells Total Total protein activity

Fraction

mg

2479 Cytosol Ammonium sulfate 10.3 1095 DEAE-cellulose 3.25 46.7 Phenyl-5PW 1.59 3.95 Mono Q 0.47 0.17

pmol AMPI

Purification pmol AMPI

-fold

min nag 0.003 0.009 0.067 0.404 2.80

min 7.42

3 23 135 936

1

TABLEI1 Sequence of tryptic peptides of human T lymphocyte GMP synthetase Tryptic peptides ofpurified GMP synthetase were separated by reverse-phase high-performance liquid chromatographyand sequenced by Edman degradation.The sequences are represented by the standard single-letter code. X indicates that no amino acid can be assigned from the corresponding cycleof Edman degradation. Peptide

Sequence

XXNIVAGIANESK ALNQEQVIAVXIDNGFM VINAAHSFYNGTXXLPISDED IIGDTFV XANEVIGXMNLK VIEPLKDFIKDEVR XPFPGPGLAI XFITSDFMTGIPATPGNEIXV IMYDLTSKPPGTTE 500

0

1000

2s

5' probe

Bgl I

1500

2000

2500

+

pcr.2S8A SA

-

f

middle prybe

3 probe

GMPS.6

Drd I

HpaI

FIG.2. Restriction map of GMF'S.6 and probes. The region of hatched bars indicates the position of the longest open reading frame. The scale gives length in base pairs. The black bar marks the length and position of the PCR fragment pcr.2S8A. The open bars mark the length and position of the cDNA fragments used as probes in Northern and Southern hybridization analysis. The arrows mark the position and direction of oligonucleotideprimers 2s and M i n the 5' to 3' orientation.

40.8% identicaltothe Dictyostelium and E. coli enzymes, respectively (Fig. 4A). The sequence of GMPS.6 also shares sequence similarity with otherG-type glutamine amidotransferases with thehighest degree of similarity spanningamino acids 69-118 and 176197 of GMPS.6 (Fig. 4B).These sequences may be involved in glutamine binding and amide transfer. A conserved cysteine residue is thought to be involved in catalysis by forming a glutamyl y-thioester intermediate (42-45). Functional Expression of GMP Synthetase-In order t o determine if the GMPS.6 clone was functional, we tested it for AT2465 activity by complementation of a guaA mutation in the strain of E. coli. This strain lacksa functional GMP synthetase and does not grow in minimal medium unless guanosine is shown). An expression vector was present (Ref. 31 and data not constructed by subcloning the sequence spanning bases1-2090 of GMPS.6 into pTRC.99A, which is a prokaryotic expression vector that contains an ampicillin-resistance marker and the trc promoter, which is inducible by IPTG. This plasmid was designated NF.6 and contained the entire open reading frame and allof the 3'- butnone of the 5'-untranslatedsequences. As a negativecontrol, NF.6 was modified tocontain two stop

23833

codons 39 bases downstreamfrom the start site of translation. This plasmid was designated ES.6 and should produce a truncated protein.Both ES.6 and NF.6 were transformedinto AT2465 E. coli, and the cells were allowed to grow on minimal medium plates containing ampicillin and IPTG with or without guanosine. Cells that were not transformed did not produce colonies on either plate due to the absence of ampicillin resistance in the parental AT2465 cells (not shown). Fig. 5 shows that while both NF.6- or ES.6-transformed cells produced colonies in the presence of guanosine, only NF.6-transformed cells formed colonies in the absence of guanosine. This result showed that GMPS.6 contained a cDNA that complements an E. coli guaA mutation. We next examined the transformed cells for the presence of GMP synthetase enzyme activity. Fig. 5 shows that GMP synthetase activity was present in NF.6 transformed cells and absent in ES.6 transformed cells. Furthermore, the activity was induced by the presence of IPTG. There was no detectable GMP synthetase activity in untransformed cells (not shown). These findings confirm the complementation results and demonstrate that we have cloned a cDNA encoding a functional human GMP synthetase. Analysis of GMP Synthetase mRNA-GMP synthetase message in A3.01 cells was examined by RNA hybridization analysis. To determine whether there were multiple messages for GMP synthetase, we probed the RNA blots with threedifferent cDNAfragments covering the 5' end (5'-StyI),middle (HindIIISac I) and 3' end (BglII-3') of the coding region (Fig. 2). All three probes hybridized to a single messageof 2.4 kb in samples containing poly(A)-selected RNA (Fig. 6A and data notshown). From these studies, we concluded that therewas a single message 2.4 kb in lengthcoding for GMP synthetase inA3.01 cells. Using theHindIII-Sac1 probe, we then examined a variety of human cell lines for the presence of GMP synthetase mRNA (Fig. 6B). All cell types examined showed a single message of 2.4 kb for GMP synthetase, the same as inA3.01 cells. However, the level of message varied dramaticallybetween transformed and nontransformedcells. All ofthe leukemic and lymphoid cell lines appeared to have similar levels of message (Fig. 6B,lanes 4-9). The level of expression of these lines was significantly higher when compared with the two nontransformed cell lines (Fig. 6B, lanes 1 and2) andperipheral T-lymphocytes (Fig. 6B, lane 3 ) . In another experiment, we examined a variety of human tissues for GMP synthetase message. Similar to thecell lines, we found a single 2.4-kb message in all tissuesexamined (data is one not shown). In summary, these results suggest that there message for GMP synthetase in humancells. Analysis of Genomic DNA for GMP Synthetase-We next examined human genomic DNA t o ascertain whether there may be more than one gene coding for GMP synthetase. DNA from A3.01 cells was examined by Southern hybridization analysis using the HindIII-Sac1 probe. Fourteen different restriction endonucleases were used and many gave single bands that hybridized with the probe (data not shown). These results are consistent witha single genefor GMP synthetase in the human genome. Alteration of Expression Level of GMP Synthetase-The elevated level of GMP synthetase mRNA in the transformed cells compared with the nontransformed cells suggests that an increase in the activity of GMP synthetase is required for increased levels of metabolic activity associated with proliferation. It can be proposed thatwhen cells are continuously proliferating, as with transformed cells, the requirement for guanine nucleotides is high, hence the expression of enzyme is elevated. Conversely, when proliferation is arrested, the re-

23834

Alteration ofExpression by Inhibition of Proliferation -LO GAATTCGGCACGAGCTCCTCTCCGC~;CCGGCTGCTGC~CTCGACCAGGCCTCCTTCTCAACCTCAGCCC GCGGCGCCGACCCTTCCGGCACCCTCCCGCCCCGTCGCCGCGGCTCCGGCCCTGGCCCCG

ATG GCT CTG TGC AAC GGA M A L C N G

1

l 61

1

CAC CAC CAC TAT GAA GGA l H H H Y E

G

GCT GTT GTC ATT CTG A V V I

ATA GAC CGA AGA GTG AGG GAA 1 T D R R V R

121

4

GCA TTT GCT ATA AAG GAA 1 A F A I K

181

6

GTG TAT GCT l V Y

241

8

L

o

ATC TCT GGA GGA CCT AAT TCT I S G G P N S

I

TGG TTT GAT CCA GCA ATA TTC ACT ATT GGC AAG CCTGTT P W F D P A I F T I G K P V

I

K

S

V

R

GAT GGA GTTTTC E D O V

AAC ATTAGT F N I

Z

421

AGG GGC C T T CAG AAG GAA GAA GTTGTTTTGCTT ~ ~ R G L Q K E E V V

1

GAT GCT GGT GCT CAG TAC GGG AAA GTC D A G A Q Y G K V

CAA GGA TTC CGTGCTATTATC E Q G F R A I

GAA GAT GCT CCC A E D A

AAA AGT GTC AGA GAA

161

181

CTT AAG GAT GGC L K D G

D

CTG TTC GTG CAG TCT GAA ATTTTC CCC TTG GAA ACA CCA E L F V Q S E I F P L E T P

L I

GGA GGA GAC G G

CTT GGA ATT TGC TAT GGT ATG CAG ATG ATG AAT AAG GTA T T T GGA GGT ACT GTG CAC AAA i L G I C Y G M Q M M N K V F G G T V H K

301

1

GAC TCC AAG CTG GAG AAT GCT D S K L E N A

GTG GAT AAT ACA TGTTCATTATTC S V D N T C S L

ACA CAT GGA GATAGTGTA L L T H G D

F

GAC AAA GTA S V D

K

V

GCT GAT GGA TTC AAG GTT GTG GCA CGT TCT GGA AAC ATA GTA GCA GGC ATA GCA AAT GAA 6 1 A D G F K V V A R S G N I V A G I A N E

(1) 541

GGA GCA CAG TTC CAC CCT GAA GTT GGC CTT ACA GAA AAT GGA AAA Y G A Q F H P E V G L T E N G K

TCT AAA AAG TTATAT i S K K L

1

8

601

GTA ATA CTG AAG AATTTCCTTTAT I V I L K N F

L

AAC AGA GAA ~ I N

CTT GAG TGTATT R E L E

CGA GAG ATC AAA GAG AGA GTA GGC ACG C I R E I K E R V

GTTTTACTC 1 V L

AGT GGT GGA GTA GAC TCA ACA GTTTGT ACA GCTTTGCTAAATCGTGCTTTG L S G G V D S T V C T A L L N R

~ 661

~ 721

x

GAT ATA GCT GGA TGC AGT GGA ACC TTC ACC GTG CAG Y D T A G C S G T F T V TCA AAA GTT TTG G T S K A (2)

Q V

L

P

E

L

2s '81

2

W C CAA GAA CAA GTC ATT GCT G?JG CAC ATT GAT AAT GGC TTT ATG AGA AAA CGA GAA AGC s l N Q E Q V I A V X I D N G F M R K R E S CAG T C TG T T GAA GAG GCCCTC 1 Q S V E E A

841

FIG.3. Nucleotideandpredicted amino acid sequences of human GMP synthetase.Amino acidresidues are represented in single-letter code. Lines and numbers inparentheses mark the position of sequences corresponding to the tryptic peptides determinedfrom the purified native enzyme. Boxes mark the position of sequences corresponding to the oligonucleotide primers. Nucleotide residue 1 represents the start site of translation.

AAA AAG C T T GGA ATT CAG GTC AAA GTG ATA AATGCTGCT

2

8

901

CAT TCTTTC TAC AAT GGA ACA ACA ACC CTA CCA ATA TCA GAT GAA GAT AGA ACC CCA CGG 0 1 H S F Y N G T T T L P I S D E D R T P R

L

K

K

L

G

I

Q

V

K

V

I

N

A

A

(3) 3 961

) 1021

~ 1081

3 1141

3

GGG GAT ACT TTTGTT ~ I G D T

GAA GTA ATT GGA GAA A N E V I

GAG GTTTTCCTT 6 1 E V

CGG CCT GAT CTA L R P D

AAG ATT GCCAAT P V K I (5) GCC CAA GGT ACT TTA F L A Q G T

GCA AGT GGC AAA GCT GAA CTCATC 9 1 A S G K A E L

1201

4

AAA ATCATT

AAA AGA ATT AGC AAA ACG TTAAAT ATG ACC ACA AGT CCT GAA GAG AAA AGA u K R T S K T L N M T T S P E E K

o

TTG AGA GAG GAG GGA l L R E

AAA ACC CAT CAC

I

K

T

H

R

K I I (4) ATG AAC TTG AAA CCA GAG G E M N L K

ATT GAA AGT GCA TCCCTTGTT L I E S A S L V

AAT GAC ACA GAG CTC ATC AGA AAG H N D T E L I R K

AAA GTA ATA GAA CCT CTG AAA GAT TTT CAT AAA GAT GAA GTG AGA

E

G

K

V

I

E

P

L

K

D

F

H

K

D

E

V

R

(6)

1261ATTTTG ~ 1 1

I

GGC AGA GAA L G

R

CTT GGA CTT CCA GAA GAG TTAGTTTCC E L G L P E E L

V

AGG CAT CCA TTT CCA GGT S R H P F P

G

(7) 1321

~ 1381

4 1411

4 1501

~ 1561

~ 1621

5 1681

5

CCT GGC CTGGCA 4 1 P G

GAA ACC AAC AATATTTTG 6 1 E T N N ACC CTATTA 8 i T L

1861

6 1921

6

GAA TCA CTTATTTTT 4 1 E S L I

L

K

I

V

A

D

F

E

GAT CAG GAG AAG D Q E

K

CTG ATG CAA L M Q Q

AGT TAC GTG TGT GGA ATC TCC AGT AAA GAT GAA CCT GAC TGG S Y S Y V C G I S S K D E P D

W

CTG GCT AGG CTT ATA CCT CGC ATG TGT CAC AAC GTT AAC AGA GTT F L A R L I P R M C H N V N R V ACA GAT GTT ACTCCCACTTTCTTG P T D V T P T

F

L

ACA ACA GGG GTG CTC AGT ACTTTA CGC CAA GCT GAT T T T GAG GCC CAT AAC ATTCTC AGG l T T G V L S T L R Q A D P E A H N I L R GAG TCT GGG TAT GCT GGG AAA ATC AGC CAG D l E S G Y A G K I S

ATG CCG GTG ATTTTG ACA CCA TTACAT TTT Q M P V I L T P L H F

GAT CGG GAC CCA CTT CAA AAG CAG CCT TCA TGC CAG AGA TCT GTG GTTATT CGA ACC T T T l D R D P L Q K Q P S C Q R S V V I R T . . "" ATTACT AGT PACTTC ATGACT GGT ATA CCT G q ACA CCT GGC AAT GAG ATC CCT GTA GAG 4 1 1 T S D F M T G I P A T P G N E I P V E l

6

GTG GTATTA AAG ATGGTCACT 1 V V L K M

V

GAG ATT AAG AAG T E I K

ATTCCT K I

GGT ATTTCT CGA ATT ATGTAT P G I S R I M Y (91

2041

6

P

GCA AGT GTT AAA AAG CCA CAT S A S V K K P H

GGT GTG CAG V G V

GTTTAT ATA T T T GGC CCA CCA GTT AAA GAA CCTCCT 6 1 V Y I F G P P V K E P

1981

6

I

CAG AGA GTC AAA GCC TGC ACA ACA GAA GAG L Q R V K A C T T E

GGT GAC TGTCGTTCCTAC 2 1 G D C R

M 6

AAA ATA GTA GCT GAT TTTTCT

AAG GAC T T TC C T I C K D F

ATT ACC AGT CTG CAT TCA CTG AAT GCC TTC TTG CTG CCA ATT AAA ACT GTA o i I T S L H S L N A F L L P I K T

1741

lSDl

ATC AGA GTA ATA TGTGCT GAA GAA CCTTATATTTGT L A I R V I C A B E P Y

8

GAC TTA ACA TCA AAG CCC CCA GGA ACT ACT GAG TGG GAG TAATAAACTTC 1 D L T S K P P G T T E W E

~

Alteration of Expression Inhibition by

of Proliferation

23835

A

1 60 MALCNGD Human . . . . . . . . . . . . . QKLENAGmL KDGHHHYEGA . .G.-W AQYGWIDRR D.discoideum MTITSPVIKT PPLNSEIRLE BNLTVESoDI EINEKDIVNA SEVIVILDAG SQYSKVIDFIR M TENIHK. . . . .HRIL=DFQ SQYTQLVW E.coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

61 ...... HumanVRELFVOSEIFPLETPAFAIKEQGFRAIIISGQP Nm-DUW pDPAILT..I D.discoideum V R E L N V A W HPLNIDLLEL IKIKSKSGST IKGIIISGGPE m G E N u K FDKSLPSEKL . . . . . .SGUL=P EBTTEENSPR APQW-EA.. E.coli VRELGVYCEL WAWDVTEAQL RDFNP

J.

121 Human EKEPVLGXXGM Q m K m B m K K S v q E 9 mFKS . .V . D.discoideum NLE1FGDX.GM Q L W I W KVESNBQRED GVHEEILKD E.coli EVPEYFEVCG M Q T W Q L Q Q HVEASNEEF EYAQVEXVND

120

180 RGLQKE .... ENQQLVSKZK N W Q T . . . . . . . . . . SALV mIEDALTAD

. . . DNTCSLF

181 .... HGDSVDWAD G-XBGN S -. KKLEAQFHP EVGLTENGKV Human EVVLLT LGWV- m X T H K D.discoideum . . . . E Q m T HGDSVTKIAD GPRIICKBDDG n S m E u R E.coli G K P L L D W S mDI(IPTAIPS DPITV-ASTES CPFAI-E KRFNVQFHP EVTHZRQGMR

240

241 300 Human ILKNFLYDIA m G T F T V Q N RELECIREIL_K ERVATS-V LLSGOWST V CTALLNRALN D.discoideum M F S E I D I C OCSANYTLDD RgQQAITYIL_K S1-N-V LVSGGVDSTV CA&JSMIG E.coli MLERPVRDIC QCEALWTPAK IIDDAVARZREQVGDDmIL GLSGGVDSSY T A M U H m I G

FIG.4. Similaritybetween human GMP synthetase and other amidotransferases. The residues that are identical to the human GMP synthetase sequence are underlined and in boldface. The arrow marks the conserved cysteine residue.A, the predicted protein sequence of the entire coding regionof human GMP synthetase is aligned to GMP synthetase from D. discoideum and E. coli.Amino acid residue 1 represents the first methionine of D. discoideum GMP synthetase. B , the putative glutamine binding domain of human GMPsynthetase is aligned with that of other G-type amidotransferases. huGMPS, human GMP synthetase; huCTPS, human CTP synthetase (38); yeAS, yeast-anthranilate synthase (39); ecPABS, E. coli para-amino-benzoate synthase (40); haCPS, hamster carbamoylphosphate synthetase (41).

301 Human QEQVIAWID NOPMRKRESQ S = W K K . L GIQVKVINU HSFYNGTTTL PISDEDRTPR D.discoideum PINVIALHID N”MIWDESL mK&SV.L QLHLIDrnS QTFYNSTTTI . . . . . . . . . K ............... E.coli .KNLTCYFVD NGLLRLNEAE QyLDMFGDHF ELNIVHVPqE .DREL

360

361 Human KRISKTLNMT TSPEEKRKIIO D T F V K I W V I G E M N m EEV.FWOGTL LD.discoideum GHLTSSLKET ISPEERMI1 GDTEMRVAEN EVKKLGLQm DV.YWOGTL RPDLIWSBK E.coli . . . . S W G E NDPEAKRKII GRVFVEVFDIE....ALKLE D y K W U T I YPDVIESAA.

420

421 480 Human VASGKAELIK THIINDTELIR KLBEEQKVII PLKDFHKDEV _RILDRE&GLg EELVSRAPFP D.discoideum T V W W V I K THHNDTELVR I L B D S E R W PLRDYHKDEV REUKSLOLS DSLWQQPIP E.coli SAT-VB S-GGLPK EMKMG..LVE_ PLKELFRDEV RKIGLELGLP YDMLYRAPFP 481 Human GPGWIRVIC AEKYICKDFPETNNILKIV ADFSMVKKP H T L L Q R W C TTEEDQEKLM D.discoideum GPGWIRIIC W E Y L V . N Y DF-QYL VTGEABSELE SEVKIKIDKQ LTEMKCKRQD E.co1i -V??LG EVKKEYCDLL RRADAIF . . . . . . . IEELRK ADLYDKVSQAFT . . . . . . . . 541 600 Human QDSLHSLNA FLLPIKTVGV PEDES= CGI . . . . .PS KDEPDWESLI FUSLIPRMC D.discoideum KD....IKP V L L P I O W QODGETYgYL LELYSSENST IDQIPHSYIF NmTIPKIE M m W X DVSL W ........................... E.coli . . . . . . . . . . V F U V R S W 660 601 Human HNVNRYVYIF GPPV . . . . . . . . . W P T D V TPTFLTTGmSTLRQmFEA HNIhRESGYA D.discoideum m I M V V F I F SQNATKHTNI K V S m V K H I TPTRLTPDVI K Q L Q W S I V SEQLYKYNLI E.coli . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

661 Human G K I m M B I L TPLHFDRDPL QKQPSCQMV VIRTFITSEm I m T E P N EIPVEWLKM D.discoideum KSLQPVPWS LPIDFG . . . . . . .VTGNWI AIRTFITNDF MTGVPAIEK NISFDCLQEI E.co1i . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M V E T I E = M A . . . . HLPYDFLGRV 749

540

720

721 Human VTEI.KKIPE E R I M y D L m KPPGTTEwe D.discoideum TNNILTSVNE ISKVLFDCTS KPPGTTEFL E.coli SNRLINEVNG XRVVYYPISG KppAZIEwg

B huGMPS huCTPS yeAS eCPABS

hams

huGMPS huCTPS yeAS ecPABS haCPS

69 364 56 44 59

F R A I I I S G G P N S W A E _A DP W F D P A I FITD K . . . P V LIGC Y G M Q WV K” I T J AHGVLVP E F G V R G T E GK I . .. Q U A WA F 3 J Q K W F EV C L m L A WE F S R N y P D T L L IPs E G H P K T D G S ISWIRYF TOKIP..WG V C L Q H Q W Q A F g ’ K y PQKZVXS PGECTPD.EA GISLDVIRHY AERLP..IE VCLDHQWQ AMKy Y D G L F LNSo P G D P A S YGPW A T L N R VSLE P N P R P pXEL G H Q L L AALI E A K T

176 512

OI ANESKKLYEAQFHPEVGLTE

118 413 106 93 111

197

VELEDHPFFVGV Q Y m F L S R P 533 1 6 7 GV RHKKYTVEQV QFBPESILTE 188 1 5 4 OI RHRQWDLEGV QFBPESILSE 1 7 5 1 6 7 OI VHDSLPFFSVQFHPEHRAGP 188

quirement for guanine nucleotides is lowered, and the expres- these cells to adhere to plastic surfaces, stop dividing, and possibility differentiate into macrophage-like cells (46-48). sion of enzyme is reduced. We were interested in the that the expression of GMP synthetase would be down-reguU937 and HL60 cells were treated with TPA for 3 days. lated if transformed cells were induced to change their meta- During this time, the majorityof the cells had adhered to the bolic state. To investigate theeffect of arresting proliferation on tissue culture flasks and had not proliferated since the start of GMP synthetaseexpression, we examined U937 and HL60 treatment, indicating that growth had been arrested. Vehiclecells. Both of these cell lines undergo differentiationfollowing treated control cells continued t o divide and remained in sustreatment by phorbol esters. Phorbol esters are known to causepension during the treatment. After the third day, the differ-

23836

Alteration of Expression by Inhibition of Proliferation a. NF.6 transformed

b. ES.6 transformed

*

1

2

9.0 E

3

c E

28S+

-

3

C2.4 kb

E

18S+ 0.0

IF’TGduring growth

+

-

+

-

Colony formation (plus guanosine):

Yes

Yes

Colony formation (minus guanosine):

Yes

No

FIG.5. Functional expression of GMPS.6 in a GMP synthetasedeficient strain of E . coLi. The expression vectors NF.6 and ES.6 were transformed intoAT2465 E. coli, and the cultures were spread on minimal medium plates with or without guanosine; colony formation was monitored after 24 h. To determine GMP synthetase activity, single colonies were selected, propagated, and induced with IPTG. Single colonieswereselected from “minusguanosine”plateswithNF.6-transformed cells and from “plus guanosine” plates with ES.6-transformed cells. Cultures were harvested, and the cytosolic fractions were assayed for GMP synthetase activity. All procedures were carried out as described under “Experimental Procedures.”

entiated adhered cells (TPA-treated) and the undifferentiated nonadhered cells (vehicle-treated)were harvested. Total RNA was prepared and examined for the presence of GMP synthetase mRNA (Fig. 7A). The TPA treatement dramatically reduced the levels of GMP synthetase mRNA where thelevels in the differentiated U937 and HL60 cells were 14 and3%of those in the proliferating cells, repectively. Cell extracts from the same treatmentgroups were examinedfor the presence fo GMP synthetase protein by immunoblot analysis (Fig. 7B). Similar to the mRNA, the level of GMP synthetase protein in HL60 cells following TPA treatment was reduced to 28% of the untreated control cells. These resultsshow that theexpression of GMP synthetase is sensitive tocell proliferation and that the arrest of proliferation causes a decrease in enzyme expression at both the mRNA and protein levels. DISCUSSION

Our dataprovide irrefutable evidence that we have obtained a cDNA that encodes a functional human GMP synthetase protein. Since obtaining the data presented here, we have successfully expressed the cDNA in baculovirus-infected insect cells. The recombinant enzyme has been purified and characterized and has identical kinetic and biochemical properties as the native human enzyme.3 Cloning of IMP dehydrogenase cDNA lead to the discovery that there are two separate genes in human coding for IMP dehydrogenase (type I and type 11) (22). The two isoforms are identical in sizes and similar in theirsequences and biochemical properties (20). Unlike the data for IMP dehydrogenase, RNA and DNA hybridization experiments indicate a single message and one gene for human GMP synthetase (Fig. 6 and data not shown). However, these findings do not preclude the existence of isoforms of GMP synthetase createdby posttranslational modification. Although the two isoenzymes of IMP des L. Lou, J . Nakamura, s.Tsing, B. Nguyen, K. Straub, H. Chan, and J. Barnett, unpublished results.

5s+

B -2.4 1

2

3

4

5

6

7

8

kb

9

FIG. 6.RNA hybridization analysis of GMP synthetase mRNA in human cells. A, A3.01 cells. RNA was isolatedfrom A3.01 cells, and Northern hybridization analysis was performed with the HindIII-Sac1 probe, a s described under “Experimental Procedures.” Lane I , total RNA(l3 pg);lane 2, poly(A)+RNA(2 pg).B,human cell lines. Total RNA from various cell lines were probed with HindIII-Sac1 fragment. The following cell lines were examined:lane 1, HFF (human foreskinfibroblast); lane 2, W138 (human lung fibroblast);lane 3 , human peripheral T-lymphocyte; lane 4 , VI3 (human T-lymphoblastoid cell); lane 5 , UC (Epstein-Barr virus transformed human lymphoblastoid cell); lane 6, Jurkat (acuteT-leukemic cell); lane 7, HL60 (human premyelocytic cell); lane 8, CEM (human T-lymphoblastoid cell); lune 9, A3.01. Ethidium bromide staining indicated similar amountsof RNA were loaded to all lanes.

hydrogenase are similar proteins, their messages are clearly distinguishable and their expression levels are differentially regulated (20, 22, 24, 29). The type I IMP dehydrogenase expression is constitutive, and it does not vary between normal leukocytes and transformed leukemic cells, while the type I1 expression is elevated in leukemic cells. Furthermore, type I1 mRNA is down-regulated in differentiated HL60 cells. Similar to thetype I1 IMP dehydrogenase, GMP synthetase expression is also elevated in transformed cells and down-regulated in differentiated HL60 cells (Figs. 6 and 7). Considering the tandem cooperation between the two enzymes, itis likely that there are common elements in the regulation of their expression. Recent studies suggest that inhibiting enzymes in the de novo synthesis pathway of purine and pyrimidine nucleotides can result in theinhibition of immune cell functions. This has proven true with the inhibition of IMP dehydrogenase by mycolphenolic acid (11, 15, 16) and the inhibition of dihydroorotic acid dehydrogenase by Brequinar (50-52). Since accumulation of guanine nucleotides is crucial for cell proliferation, it islikely that theinhibition of enzymes downstreamof IMP dehydrogenase would also be effective in inhibiting immunecell functions. GMP synthetase activity is elevated in many tumor tissues and

Alteration of Expression by Inhibition Proliferation of

23837

4. Moldave, K. (1985)Annu. Rev. Biockem. 54, 1109-1149 5. Hepler, J. R., and Gilman, A. G. (1992)Trends Biockem. Sci. 17,383-387 6. Weher, G. (1983)Cancer Res. 43, 3466-3492 '2.4 kb 7. Weber, G., Nakamura, H., Natsumeda, Y., Szekeres, T., and Nagai, M. (1992) Adu. Enzyme Re&. 32, 57-69 8. Hirai, K., Matsuda, Y., and Nakagawa, H. (1987)J. Biockem. (Tohyo) 102, 893-902 9. Kiguchi, K., Collart, F.R., Henning-Chuhh, C., and Huherman, E. (1990)Cell Grotutk & Diffe'fer.6, 259-270 1 2 3 4 5 6 10. Yu, J., Lemas, V., Page, Y., Connor, J. D., and Yu, A. L. (1989)Cancer Res. 49, 5555-5560 " " " 11. Mitchell. B. S..Davton. J. S.. Turka. L. A.. and Thommon. C. B. (1993)Ann. N. E:Acad. Sci.bS5, 217-224 12. Dayton, J. S., Turka, L. A,, Thompson, C. B., and Mitchell, B. S.(1992)Mol. Pkarmcol. 41, 671676 13. Hasunuma, T., Yamanaka, H., Terai, C., Miyasaka, N., Kamatani, N., Nishioka, K., and Mikanagi, K. (1989)Adu. Exp. Med. Biol. 2534 455-460 14. Nichols, K. E., Chitneni, S. R., Moore, J. O., and Weinherg, .I. B. (1989)Blood 74, 1728-1737 15. E u y i , E. M., Almquist, S. J., Muller, C., D., and Allison, A. C. (1991)Scand. J. Immunol. 33, 161-173 16. E u y i , E. M., and Allison, A. C. (1993)Ann. N.Y.Acod. Sci. 685,309-329 'GMP 17. Ensley. R.D., Bristow, M.R., Olsen, S. L., Taylor, D. O., Hammond, E. H., synthetase OConnell. J. B., Dunn, D., Osburn, L., Jones, K. W., Kauffman, R. S., Gay, W. A,, and Renlund, D. G . (1993)?Fansplanation 56, 75-82 18. Goldhlum, R. (1993)Clin. Exp. Rkcumatol. 11, Suppl 8 ; S117-119 19. Sollinger, H. W., Belzer, F. O., Deierhoi, M. H., Diethelm, A. G., Gonwa, T. A., Kauffman, R. S., Klintmalm, G . B., McDiarmid, S. V., Roberts, J., Rosental, J. T., and Tomlanovich, S. J. (1992)Ann.Strrg. 216,513-518 20. Carr, S.F.. Papp, E., Wu, J. C., and Natsumeda, Y. (19931J. Biol. Ckem. 268, 2728627290 21. Antonino, L., and Wu, J. C. (1994)Biochemistry, 33, 1753-1759 22. Natsumeda, Y., Ohno, S., Kawasaki, H., Konno, Y., Weber, G., and Suzuki, K. (19901J. Biol. Ckem. 265,5292-5295 Hodges, S. D., Fung, E., McKay, D. J., Renaux. B. S . , and Synder, F.F. (1989) FIG.7. The effect of TPA on t h e e x p r e s s i o nof GMP synthetase 23. J. Bid. Ckem. 264, 18137-18141 in HL60 and U937 cells. HL60 or U937 cells were treated with 0 24. Nagai, M., Natsumeda, Y.,and Weber, G. (1992)Cancer Res. 52, 258-261 (control) or 20 ngiml TPA for three days. Nonadhered controlcells and 25. Tiedeman, A. A,, Smith, J. M., and Zalkin, H. (1985)J . Biol. Ckem. 260, adhered TPA-treated cells were harvested and analyzed. TheRNA and 86768679 protein bands were quantitatedby densitometry. A, RNA was isolated 26. MlintsPIti, P., and Zalkin, H. (1992)J. Bacteriol. 174, 1883-1890 and analyzed by RNAhybridization analysis.Lunes 1-6 represent U937 27. Van Lookeren Campagne, M. M., Franke, cJ., and Kessin, R. H. (1991)J.Biol. Ckem. 266, 16448-16452 total RNA, lunes 7-12 represent HL60 total RNA, and lune 13 representsA3.01poly(A)' RNA. Lanes 1 3 and 7-9 represent control samples, 28. Dujardin, G., Kermorgant, M., Slonimski, P. P., and Boucherie, H. (1994)Gene (Amst.) 139, 127-132 and lunes 4-6 and 10-12 represent TPA-treated samples. Thefollowing 29. Folks, T., Benn, S., Rahson, A,. Theodore, T., Hoggan, M. D., Martin, M., amounts of RNA were loaded: 5 pg inlunes 1,4,7, and 10;10 pg inlunes Lightfoote, M., and Sell, K. (1985)Proc. Nall. Acad. Sci. U. S. A. 82,45392,5,8, and 11; 20 pg in lanes 3 , 6 , 9 , and 12; and 1.2 pg in lune 13. €3, 4543 immunoblot analysis of HL60 cell lysate. Lysates of HL60 cells were 30. Eugui, E. M., and Almquist, S.J. (1990)Proc. Natl. Acnd. Sei. U. S. A. 87, immunoprecipitated with either preimmune antisera (lanes I , 3, and 5 ) 1305- 1309 or anti-GMP synthetase antisera (lunes 2, 4 , and 6). The precipitated 31. Taylor, A. L., and Trotter, C. D. (1967)Bacteriol. Rev. 31,332-353 proteins were analyzedby immunoblot analysis with anti-GMP synthe- 32. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J.. and Rutter, W. J. (1979) tase antisera.Lanes 1 and 2, control HL60; lunes 3 and 4, TPA-treated Biochemistry 18, 5294-5299 33. Aviv, H., and Leder, P. (1972)Proc. Noll. Acad. Sei. U. S. A. 69, 1408-1412 HL60; lunes 5 and 6, A3.01. 34. Samhrook, J., Fritsch, E. F., and Maniatis, T.(1989) Molecular Cloning: A LaboratoryManual, pp. 7.3-7.83,Cold Spring Harbor Laboratory, Cold rapidly proliferating cells (6-8). Here we demonstrate that the Spring Harbor, NY expression of mRNA is dramatically elevated in transformed 35. Spector, T. (1978)Methods Enzynlol 51, 219-224 36. Heller, R. A,, Song, K., Onasch, M. A,, Fischer, W. H., Chang, D., and Ringold, cell lines compared with nontransformed cell lines (Fig. 6B). G. M. (1990)Pror. Natl. Acad. Sei. U. S. A. 87, 6151-6155 This suggests that the increaseenzymatic in activityis a result 37. Boritzki, T. J., Jackson, R. C., Morris, H.P., and Weber, G. (1981)Biockim. Biopkys. Acta 658, 102-110 of increased protein expression.Furthermore, inhibition of pro38. Yamauchi, M., Yamauchi, K., and Meuth, M. (1990)EMBO J. 9,2095-2099 liferation causes down-regulation of GMP synthetase expres- 39. Zalkin, H., Paluh, J. L., van Cleemput, M., Moye, W. S., andyanofsky, C. (1984) J. Biol. Ckem. 259,3985-3992 sion a t both the mRNA and protein levels (Fig. 7). Therefore, it Kaplan, J. B., and Nichols, B. P.(1983),I. Mol. Bid. 168, 451-458 seems apparent from the collective data that GMP synthetase 40. 41. Simmer, J. P.,Kelly, R. E., Rinker, A. G., Jr., Scully, J. L., and Evans, D. R. would likely be a successful target for cancer chemotherapyand (1990)J. Biol. Ckem. 265, 10395-10402 immunosuppression. For this reason, we are currentlyinvesti- 42. Levitzki, A., and Koshland, D. E.,Jr. (1971)Biochemistry 10, 3365-3371 43. Mizohuchi, K., and Buchanan, J. M. (1968)J. B i d . Ckem. 243,4853-4862 gating the regulationof GMP synthetase geneexpression and 44 Weng, M., Makaroff, C. A,, and Zalkin, H. (1986)J. Biol. Ckem. 261, 5568its role in cell growth and development. 5574 45 Tso.. J. Y.., Bower., S. G.., and Zalkin. H. (1980)J. Biol. Ckem. 255.6734-6738 46 Rovera, G., Santoli, D., and Damsky, C. (1979)Proc. Natl. Acad. Sei. U. S. A. Acknowledgments-We thank Dr. Renu Heller for helpful sugges76.2779-2783 tions. We also thank Dr. Kurt Jarnagin and Natalie Saldou for sequencing the tryptic peptides and Dr. Chinh Bach and Patricia Zuppan for 47 Huberman, E., and Callahan, M. (1979)Proc. Natl. Acad. Sci. U. S. A. 76, 1293-1297 sequencing the cDNA clones. 48 Harris, P., and Ralph, P. (1985)J. Leukocyte Biol. 37, 407422 49 Deleted in proof REFERENCES 50 Chen, S. F., Ruben, R. L., and Dexter, D. L. (1986)Cancer Res. 46,5014-5019 51 Jaffee, B. D., Jones, E. A,, and Loveless, S. E., and Chen, S.E (1993)Trans1. Gelfand, V. I., and Bershadsky, A.D. (1991)Annu. Rev. Cell. Biol. 7,93-116 plant. Proc. 25, 19-22 2. Huhbard, S.C., and Ivatt, R. J. (1981)Annu. Reo. Biockem. 50,555-583 3. Stayton, M. M., Rudolph, F. B., and Fromm, H. J. (1983)Cum Top. Cell. Regul. 52 Simon, P., Townsend, R. M., Harris, R. R., Jones, E. A,, and Jaffee.B. D. (1993) 22, 103-141 Transplant. Proc. 25, 77-80

1 2 3 4 5 6 7 8 9 101112

13

- -~"-w ~

d

~I

~